Improving spending efficiency to maximize results

State and local governments are entering challenging fiscal times. This will require hard decisions about funding priorities and the use of data and evidence to preserve the most important services. 

However, governments also have an opportunity to improve the efficiency of their programs to maximize the impact of taxpayer dollars with technological advances paving the way for this progress.

While the word “government efficiency” has become politicized, there are a number of common sense and nonpolitical ways that governments can operate more efficiently. One area that is particularly ripe for revamp is the government contracting and grant-making process, which still often uses manual processes and outdated technology.

For context, in 2024 the federal government spent $1.2 trillion on grants, mainly for non-federal organizations to provide services on behalf of federal departments and agencies. A large proportion of these funds flow to state and local governments who in turn re-grant a significant amount of these dollars to organizations in their community. 

These organizations then provide services to local residents in housing, health, workforce training, and other critical areas. For example, using funds that the U.S. Treasury Department provided to address the economic and health impacts of COVID, Ohio provided $15 million to Habitat for Humanity to develop workforce housing.

Though these federal dollars support vital services in communities, in many cases the people receiving these services do not know who is funding them or the process by which these funds reach their trusted local organizations. Nonetheless, while research shows there are relatively high levels of trust and satisfaction with government services, nearly any organization receiving government funds would agree that the grants and contracting process could be improved, with better tech tools playing a key role.

For example, government requests for proposals routinely run to nearly 100 pages, contain dense text that is written in legalese, require an excessive number of documents to be submitted in various systems, and lack clarity about how results will be measured. However, there are some basic fixes that governments can make to improve both the efficiency and impact of their grants and contracting process, maximizing impact and reducing spending.   

First, governments can streamline RFPs to focus on the most important activities and eliminate ancillary information related to legislative authority and legal requirements. This information is generally not useful to applicants for government funds. 

By moving away from static PDFs and utilizing more dynamic web platforms, governments can make this background information available to prospective bidders on an as needed basis. For example, the U.S. Department of Justice moved background information about the grant applications onto its website. Similar changes at the U.S. Department of Health and Human Services helped reduce the page count of the Department’s grant announcements by nearly 20%.

Second, governments can use artificial intelligence and other tech tools to simplify their writing. RFPs are often written with complex language, jargon and at a grade level that’s inappropriate for potential applicants. When more complex and difficult writing is used in an RFP, it makes it more difficult for people to understand and apply for it. As a result, many organizations need to pay grant writers to help them decipher government jargon. 

As a rule of thumb, if a grant writer is required then the application probably needs to be simplified. For example, a concerted approach to simplifying language allowed the Administration for Children and Families to reduce the reading level of its grant announcements by four grade levels, which allowed applicants to complete the application over 30% faster.

Simplifying writing is an ideal use case for governments’ use of AI. Even the most basic AI models can take complex text and transform it into clearer language that is more accessible for readers. And more advanced government procurement systems are helping government grants staff by building AI powered writing tools into the RFP development process itself. In addition to increasing efficiency, the use of proprietary AI tools also solves the legal challenge that governments face when they contemplate inserting non-public draft RFP language into commercial AI models.

While AI is a powerful way to streamline government writing, the ease of this process doesn’t mean everything needs to be written at a low grade level. 

Instead, governments should ensure that the writing matches the potential applicant pool for a given RFP. If issuing a grant for laboratory health research, then writing at the PhD level may be appropriate, since presumably many health researchers have actual PhD degrees. But for a grant focused on community led health outreach then writing at a lower grade level is probably the best course of action. Though in any case, even PhD researchers probably prefer clear writing.

Third, governments can improve the design of their RFPs. Using tools and technologies from the field of human-centered design, governments can structure their grant announcements so applicants can easily understand information, quickly find the most important details and have a clear sense of their next steps. Using clear headers, coloring coding, bullets and other visual cues are all important ways to make government documents — not just RFPs — easier to understand. While in past years this kind of design work might have required hours of time from a graphic designer, and slowed down the grant-making process, this work can now be automated and integrated into grants management systems, making it more accessible to governments of all sizes.

Fourth, governments should only require that applicants submit information and forms that are needed to determine the applicant organization’s capability to carry out the desired activities. For example, every RFP should require applicants to explain their qualifications for completing the grant activities. However, not every RFP should require applicants to submit a form certifying that they will take cybersecurity training if they win the grant.

This information, if it is needed at all, is the kind of pro-forma compliance information that will not help the grant maker decide which applicant is best qualified to receive funding. Rather, this information can be gathered from the winning organizations as part of finalizing the contract before they receive funds. 

In other words, if a form is only needed for the administrative processing of a grant payment then it likely only needs to be collected from the 10 winners rather than the 100 applicants — instantly reducing burden for 90% of applicants. By automating grants processes governments can link application requirements to their grants scoring matrix, thereby identifying which aspects of the grant application are likely superfluous.

Fifth, governments should make sure their grants focus on a clear set of activities that are likely to achieve the desired outcomes. By zeroing in on the most important strategies, ideally those that are evidence-based, governments can make RFPs clearer and shorter. 

For example, the Delta Regional Authority revised a workforce grant announcement to concentrate on a clear set of actionable activities, making it easier for applicants to understand what activities they should be conducting. However, this is an area where more advanced technological tools could be developed and help governments more efficiently identify and integrate evidence-based models into their RFPs.

While some of this work to improve the grants process can be potentially time consuming — though technology can speed the adoption of new practices — governments do not need to do it all at once. In fact, in many cases it makes sense to start small, learn lessons about what works, iterate to further improve processes, and then scale.

In sum, there are a number of ways that governments can improve their grant and contract process, powered by modern technology. Specifically, governments can do this by eliminating extraneous information, writing more clearly, improving design, only asking for the most relevant forms, and focusing on a core set of activities. 

When taken together, this can cut down the page length, reduce the reading level, make key information easier to find, and reduce burden on applicants. By using AI and other technology tools to increase the efficiency of this process, state and local governments can see reduced costs, more qualified applicants for their grant funds, and better results for their residents.  

Jed Herrmann provides budget and program implementation consulting services to state and local governments. He previously served as a senior official at the U.S. Treasury Department and the White House Office of Management and Budget, designing and implementing trillions of dollars of programs focused on state and local governments.